THE EVOLUTION OF FUNCTIONAL LEFT PERIPHERIES IN HUNGARIAN SYNTAX

This is the first English language book on Hungarian diachronic syntax in the generative framework. It summarizes the outstanding achievements of a four-year OTKA (Hungarian Scientific Research Fund) project on diachronic syntax of Hungarian, carried out at the Research Institute for Linguistics under the leadership of Katalin É. Kiss. The book is a somewhat more concise version of the Hungarian publication É. Kiss 2014, adapted to an international readership. The book provides an empirical overview of the syntactic changes in Hungarian, covering the following five major areas: basic sentence structure, DP structure, quantification, PP structure, and subordination. The book’s title is slightly misleading since the volume is not explicitly about the functional left periphery in all of the above areas, but rather about the development of functional material (in the left periphery or elsewhere) that is closely linked to historical change and grammaticalization. The diachronic data studied are texts originating between the end of the twelfth century and the end of the sixteenth century, from the second half of the Old Hungarian period (which roughly corresponds to six centuries between c. 900 and 1500) and the beginning of the Middle Hungarian era that followed it. In this period, the most important documents comprise forty-seven hand-written codices (Old Hungarian) and printed books in various genres (Middle Hungarian). These texts were annotated and entered into a corpus built for the project (currently available for the Hungarian-speaking audience at http://oldhungariancorpus.nytud.hu). The building of the corpus itself was a massive undertaking, the details of which (the process of digitalization, problems of character recognition, text encoding, annotation, etc.) are documented by Eszter Simon in the appendix to the volume. The five chapters detailing historical change in Hungarian share a methodology. The reconstruction of the Proto-Hungarian language (with no surviving documents) is based on the S-curve method (Croft 2000): new constructions first spread slowly, then fast, then slowly again, while old constructions are ousted following the reverse pattern. Using the backward extension of these curves, as well as comparisons with the Ob-Ugric sister languages Khanti and Mansi, the research established several features of Proto-Hungarian and Old/Middle Hungarian. The following gives a brief summary of the chapter contents. Ch. 1 by Katalin É. Kiss argues that Hungarian has changed from an SOV to an SVO language, a change that went hand in hand with the disappearance of unmarked objects and the appearance of the modern Hungarian left periphery. Possibly, rightward-dislocated phrases came to be reanalyzed as base-generated arguments, bringing about the reanalysis of preverbal elements as A-bar constituents, resulting in a Topic–Focus–V–X order (rather than, strictly speaking, SVO). The change must have taken place before the start of the Old Hungarian period since the fifty clauses of the first Old Hungarian document already utilize every clausal functional projection that is attested in Modern Hungarian. Although the head-final nature of Proto-Hungarian is argued for on the basis of evidence that VP, TP, and CP were all head-final in that period, the role of Tense is neglected throughout in the change from SOV to the Old Hungarian structure. Note that there is precious little on this topic even in the highly relevant chapter on finite and nonfinite subordination apart from an important, though rather brief, observation, claiming that ‘Old Hungarian was not an OV language any more, yet auxiliaries strictly followed the main verb, which means that VP was not head final, but TP/AspP still was’ (n. 17, p. 182). The paper also provides a detailed account of the development of object agreement, which is descended from topic-verb agreement in Proto-Hungarian, where the subject was a primary topic, and the object a focus or a secondary topic, which triggered agreement only in the latter case (an instance of differential object agreement). By the twelfth century, agreement with the primary topic generalized as subject agreement and agreement with the secondary topic as object agreement.